Semantic Kernel .NET can't call Azure OpenAI o1/o3-mini: Model o1 is enabled only for api versions 2024-12-01-preview

TL;DR

The upstream fix arrived in Azure.AI.OpenAI 2.2.0-beta1, so updating that package (and Semantic Kernel to a version that picks it up) resolves it cleanly.

Steps

1. Calling an Azure OpenAI o1 deployment (model version 2024-12-17) from Semantic Kernel .NET failed with HTTP 400: "Model o1 is enabled only for api versions 2024-12-01-preview and later". The o1 model name version is not the API version, and SK was sending an older Azure OpenAI API version. Worse, the underlying Azure.AI.OpenAI SDK had no option to select 2024-12-01-preview yet, so there was no supported way to call o1 at all.
2. The o1/o3-mini models need Azure OpenAI API version 2024-12-01-preview or later. The workaround in the thread is a custom HttpClient handler (AzureOverrideHandler) passed to the connector to override the API version (and the maxcompletiontokens limitation). The upstream fix arrived in Azure.AI.OpenAI 2.2.0-beta1, so updating that package (and Semantic Kernel to a version that picks it up) resolves it cleanly.

When to use

You are seeing this: The o1 model name version is not the API version, and SK was sending an older Azure OpenAI API version. Use this skill when you run into "Semantic Kernel .NET can't call Azure OpenAI o1/o3-mini: Model o1 is enabled only for api versions 2024-12-01-preview".

When not to use

If your error message or symptom does not match what is described above, this is probably not your fix. Search for your exact error text instead of forcing this one to fit.

Versions

Versions mentioned in the source: Azure.AI.OpenAI 2.2.0-beta1. If you are on something much newer or older, the details may have shifted.

Why this happens

The original report does not dig into a root cause. It documents the symptom and the fix that resolved it.

Question asked in the thread: Calling an Azure OpenAI o1 deployment (model version 2024-12-17) from Semantic Kernel .NET failed with HTTP 400: "Model o1 is enabled only for api versions 2024-12-01-preview and later". The o1 model name version is not the API version, and SK was sending an older Azure OpenAI API version. Worse, the underlying Azure.AI.OpenAI SDK had no option to select 2024-12-01-preview yet, so there was no supported way to call o1 at all.

Verified answer: The o1/o3-mini models need Azure OpenAI API version 2024-12-01-preview or later. The workaround in the thread is a custom HttpClient handler (AzureOverrideHandler) passed to the connector to override the API version (and the maxcompletiontokens limitation). The upstream fix arrived in Azure.AI.OpenAI 2.2.0-beta1, so updating that package (and Semantic Kernel to a version that picks it up) resolves it cleanly. Source: https://github.com/microsoft/semantic-kernel/issues/10201

Thread: https://vectle.com/threads/thr_tyJdm9Em55czKk1DFnFcIA